CoCo

As mobility is an important key to many applications, this work proposes a location-less model represent that used quantify correlations between traces collected by built-in sensors on smartphones. We analyze the from two aspects: co-direction relationship and co-movement . The former similarity of macroscopic moving directions traces, whereas latter their microscopic vibrations. To verify merits proposed metrics, exemplary use case, termed co-mobility detection , considered determine if mobile devices share same journey entity (e.g., carried person). Comprehensive experiments with diverse combinations are conducted in three different environments density Wi-Fi networks. experimental results indicate metrics can effectively evaluate both coarse-grained fine-grained movement variations along traces. accuracy algorithm achieve 90% average for duration 70 s.